Moral Panic, again on guns

A moral panic involves masses of people becoming irrational about a problem, both over-estimating how widespread and serious it is and demanding immediate action for action’s sake. I know of no good definition: Every one I have encountered provides illustrative examples that violate that definition.

Many of you will immediately think of the 1692-1693 ^ Salem Witch Trials^. Some may remember the Satanic Ritual Abuse panic of the 1980s and 1990s, with the ^ McMartin Preschool Trials^ being the most prominent example.
Quick overview: Adults manipulated the children into making bizarre claims, such as a preschool worker flying around the room, and that one of the children had been murdered in a ritual sacrifice, despite none of the children being missing. For a variety of reasons — including cynical career promotion and profit — social workers, the district attorney, and local newspapers ignored and suppressed evidence to the contrary. The LA Times in particular followed a time-honored practice of boosting circulation by whipping up outrage and fear, and when that played out, going for circulation by exposing and analyzing the panic, but of course ignoring their own role.

My goal here is not to argue for any particular actions to attempt to reduce the problem of gun violence, but rather to encourage you to take a broader and more detailed approach that what is being served up by the politicians and media. And for you to ask probing questions of the advocates to assess whether they are focused on emotional reactions or they have an appreciation of the complexities revealed by decades of trying to deal with these issues.

There are strong psychological and political effects that cause people to prioritize being seen to be doing something now, instead of doing something effective. The big danger is that ineffective measures and ^ virtue signaling^ will waste so much time, resources and momentum as to preclude what might have been effective.

Note on terminology:
– I will use “AR-15” as a shorthand for “AR-15 style rifle”.
– The term “assault weapon” appears in quotes here because it is ill-defined, intentionally misleading, and highly partisan. It deserves air/finger quotes, but those aren’t available here. The term ^ assault rifle^ has a widely accepted definition and is a different category from “assault weapon”: the former has a selector to also provide burst or automatic fire in addition to semi-automatic fire, whereas the latter includes certain semi-automatic rifles and non-rifles. The Associated Press guidelines are that an “assault weapon” is strictly semi-automatic and does not include assault rifles.
– The use of the term “mass shooting/killing” here shall exclude events such as gang-related killings and intra-family murder-suicides. It is meant to specify killings in public places involving terrorism or seeking notoriety or mental illness or where there doesn’t appear to have been a rational motive.
– The common threshold for something to qualify as a “mass killing” is 4 dead, not including the perpetrators. A “mass shooting” typically uses the same threshold, but includes both people killed and those wounded. It may also include people with non-gunshot injuries, for example, ones incurred while escaping.
Notice there can be mass killings that are not mass shootings and mass shootings that are not mass killing.
Warning: Because many variants of the terminology are used — often not specifying the definition used — it can be hard to ascertain when you are comparing apples to oranges. So be careful if you choose to delve into the statistics.

Caveat: Intuitions about rare events are often very wrong. Mass shootings are very rare when the measure isn’t the calendar, but factors such as population. When the “Action News” format was adopted by many TV stations (in the 1970s?), one slogan was “If it bleeds, it leads.” Newspapers responded by running more stories about crimes from around the country. Researchers found that this caused the readers to believe that they were now less safe than they had been (and statistically still were). Despite the TV stations and newspapers being well aware that their practices were harming society, they persisted.
Aside: The current media practice of stoking fear for profit is nothing new.

In an earlier blog, I presented a cautionary example that is commonly part of lessons about dealing with infrequent situations: There is a very rare fatal disease that has a risky cure. However, testing for the disease results in more people dying, because those who were saved were fewer than those who died from unnecessary treatment after being inaccurately identified as having the disease.(foot#1)

Act in haste, repent in leisure. Many are calling for Congress to take immediate action and pass new laws, despite Congress’ history of wretched drafting of laws. Federal judges have complained for years about having to spend inordinate amounts of time cleaning up those laws, resolving the contradictions and ambiguities, filling in the unintentional gaps …

The current push for “Red Flag Laws” — to allow the police to confiscate a person’s gun on suspicion of a problem — has the advocates ignoring the potential for deadly abuse of such laws. Maliciously false criminal accusations are very rarely punished, even when they wreak havoc and permanently damage to the life of the falsely accused. Similarly for Temporary Restraining Orders (TRO). And the courts reputedly grant TROs if there is merely a claim of a risk of violence. With a Red Flag Law the police may similarly play it safe and confiscate a gun on even flimsy and suspect reports.

As a hypothetical example, consider a woman who is being stalked and the threat is credible enough that she is granted a conceal carry permit for a pistol. Then the police confiscate her gun under a Red Flag law: Perhaps someone interpreted her stress and cautiousness as mental instability, or perhaps someone associated with the stalker made a false report. The urgency inherent in the concept of Red Flag laws largely precludes an investigation or her having a chance to give her side. By the time she has a chance to do so — weeks or months later — she may already be dead. A well-intentioned law to protect people from gun violence facilitated her murder.

“The road to hell is paved with good intentions.”
“The opposite of well done is well-intended.” (Austro-German proverb).

For the California Legislature, complicating low-risk transfers is more important than saving lives.
The California Legislature passed a law (in 2014?) to make the private transfers of guns — including temporary changes in possession (not ownership) — difficult and time-consuming. Problem: A substantial majority of gun deaths are suicides, about 70% nationally. Despite protests, the law makes it difficult for a household with a temporarily depressed person — such as postpartum depression or intense grief over the loss of a loved one — to temporarily remove a gun from the household. In response to continuing protests about this aspect, a bill intended to fix this — ^ AB-2817: Firearms: Emergency transfers (2017)^ — was drafted, but died in committee. (I mentioned this situation in an ^ earlier blog^, in the section “The inconvenient complexity of mental illness”).

The vast majority of gun homicides result from criminal gang activity: How likely is it that gang members would tell the government of transfers of illegal guns?

I have seen guns-rights advocates repeatedly claim that killings resulting from such transfers are not to be found in the statistics, although this may represent failings of the databases used. However, if such transfers were a significant problem, I would expect the gun control advocates to point it out, which I haven’t seen. Also, you might think that such data would have been gathered and presented as part of the consideration of the law, but apparently not.

Is the solution to the non-enforcement and failures of current laws to pass more of the same? Example: In the recent shooting of 6 police officers in Philadelphia, politicians and others are calling for more gun control. Wait. The shooter had multiple felony convictions, including ones for illegal possession of guns. Existing laws already make it illegal for a felon to possess a firearm. Are we going to see laws to make it illegal to illegally possess a firearm? Passing laws with no expectation of effective enforcement is just political posturing and playing to the crowd.

Could a ban on “assault weapons” be effective? I have seen numerous arguments that it wouldn’t be, and seen no credible responses to those arguments from the advocates. Why expend massive political capital to impose such a ban if you don’t know how you could make it work?

The ^ Federal Assault Weapons Ban of 1994^ was not renewed. The widely accepted assessment among researchers and others looking at the statistics is that it had little or no effect. However, you will hear prominent politicians — recently Joe Biden — and partisan commentators (bloviators?) claim otherwise.

Its goal was to ban certain models of semi-automatic rifles, and resorted to using cosmetic features, thereby creating situations where functionally equivalent variants of banned models weren’t banned.(foot#2) Interestingly, the Assault Weapons Ban did not ban assault rifles: They were already banned under the National Firearms Act of 1934.
What is the relationship of gun availability to mass shootings? Ask yourself: Could the availability of various firearms be responsible for the recent dramatic increase in mass shootings? The current “assault weapons” have been widely available for 40 years,(foot#3) semi-automatic rifles since at least the 1960s,(foot#4) and rapid-fire rifles with large magazines for over 140 years.(foot#5) The legal availability of guns is likely somewhat declining because of recent laws raising the age threshold. Ask yourself how such a decline/no-change could be responsible for an increase of mass shooting with legally obtain firearms. Rifles of all types and shotguns are involved in a tiny proportion of gun homicides (3-5%). So why is the minuscule portion of homicides that involve “assault weapons” the priority? Obvious answer: They are the ones that the media focuses on and aggressively promotes, creating a (profitable?) moral panic.

So what might be actual, actionable causes?

When is gun availability an enabler? There are many cases of individual shootings where this is the case. For example, I lived outside Detroit during the 1970s — when Detroit carried the title of “Murder Capital of the US”. In the aftermath of the 1967 Detroit riot, many residents bought guns for self and home defense. (Aside: The Detroit mayor reputedly had at least one gun stashed in almost every room.) The combination of fear and available firearms — primarily pistols — produced a stream of senseless shootings.(foot#6) With an assumption that most people were armed, there was a strong incentive to be the one who shot first. Thus, minor disputes could quickly escalate into shootings and killings.

But these types of shootings are not part of the moral panic being discussed here. Most of the mass shootings were not spur-of-the-moment but planned, sometimes for many months. If firearms weren’t available:
– How many of these would have been prevented by the lack of an adequate alternative?
– How many would have been less deadly?
– How many would have been more easily detected and prevented?
– How many would have happened but with a different mix of weapons? For example, in the ^ Columbine school shooting^, the plan was to have most of the killing done by bombs, but fortunately, none of them exploded.
I don’t have the slightest idea of the answer to these questions.

Put another way, were guns simply the weapon of convenience, the weapons of choice, or somehow essential for the killings? For example, in a workplace shooting, it may be important to the shooter to see the terror of certain of the victims. However, for a killer seeking notoriety from the number of people killed, there could be many alternatives to using guns.

When I read government and academic reports, I don’t see evidence that the authors even thought to consider these questions. My guess about the reason? “When all you have is a hammer, everything looks like a nail.” Intense specialization among researchers and their funders creates silos.

What you don’t want to have happen is to have your focus on one class of weapons, only to see the murders migrate to other weapons. Consider Britain, which has strong gun control. Yet, earlier this year, London had more murders than New York City — with knives as one of the alternatives to guns. Terrorist killings used explosives, trucks, knives, machetes …

Have “assault weapons” increased the number killed in a shooting? I couldn’t find accessible data on this, so I cobbled together a short list of the mass shootings that established new levels for people killed. What is notable from this list is that only the last — the 2016 Orlando Nightclub shooting — involved an “assault weapon”, with a slow, grossly inadequate police response greatly contributed to the number who died.(foot#7)

Should mass killings be categorized by weapon type? The assumption embedded in the term “mass shooting” is that the primary distinguishing factor should be the choice of weapon. But is it reasonable to assume that someone who kills for notoriety, someone whose mental illness causes them to lash out, and a terrorist are more alike because their weapons were guns than are, say, terrorists who used different weapons: bombs, guns, trucks, …

Mistakes in creating data sets can produce misleading and faulty results. This problem occurs routinely in machine learning. An example I recently heard of: The computer was identifying wolves in many photos in which there was nothing faintly resembling one, but they did have snow. Because too many of the photos used for training also included snow, what the computer had essentially learned was “If there is snow, there probably is a wolf.”

What is the role of the media reporting? In Palo Alto, we are painfully aware of suicide clusters and our local media has been following advice to limit coverage so as to not trigger copycats. For mass shootings, some of the national media limit reporting of the name of the shooter to discourage those who might otherwise stage their own mass shooting as their path to fame (notoriety). However, if the shooter uses an “assault weapon”, he is virtually assured of massive publicity of the event, even if his name isn’t always mentioned. Could this be encouraging the choice of these particular weapons?

Is the cost/benefit ratio of a proposal being appropriately estimated?
For example, consider the proposals to arm teachers. The costs extend well beyond the price of the guns: There are training and practice costs and the costs of safe storage containers. As a tradeoff, would you support slashing funds for instructors to help students with problems with math and reading catch up? How many of the students who would be denied help would die as an indirect result? From overdoses, crime …

How many people are going to be shot with these guns? For example, during a police officer’s visit to an elementary school, a third-grader managed to get her finger inside his safety holster, one that was supposed to prevent the pistol from being fired. It didn’t.(foot#8) Then there is human error and human malice.

Given the massive number of schools in the US, events that are low probability for an individual school become numerous when viewed nationally (see “Caveat: Intuitions about rare events are often very wrong” above). For example, if the probability of an error is one in a million, would you call it “Rare”? If 10 million events are generated every minute — by a computer … — that error occurs every 6 seconds on average.

—-What other factors are relevant?—-

A well-designed system is resilient to the inevitable failures in its components. For example, a background check can fail because someone failed to enter, or misentered, information into the database. The ^ 2009 Fort Hood shooting^ (13 killed) is but one of many examples of this.

Reports of suspicious or threatening behavior can provide another line of defense to handle failures in a background check or changes since a background check. This is protection that Red Flag laws promoted as providing. In multiple mass shootings, there have been multiple serious and urgent reports that were ignored by the authorities or deemed as not actionable. Why did authorities dismiss or ignore such reports? Is it that those reports about someone who turned out to be a mass shooter were indistinguishable from reports that go nowhere?

There is also the problem of people being reluctant to report suspicious behavior. For example, in the aftermath of the ^ 2015 San Bernardino attack^ (14 killed), neighbors were reported as having had serious concerns and suspicions about the shooters, but were afraid to report them, with the implication that they were afraid to be labeled bigots (the shooters were Pakistani Muslims).

The failure to head off some of the shootings has been attributed to the warning signs being reported to different groups, but not being shared and put together to create an actionable warning. Recognize that this is a difficult problem along many dimensions: technical, administrative, legal, privacy …

—-If you are serious about finding a real solution to mass shootings, maybe you shouldn’t persuade people that you aren’t—-

Senator Diane Feinstein (D-California) is a prime example of this. She cites the ^ 1978 SF City Hall murders of Mayor Moscone and Supervisor Harvey Milk^ as a major motivation for her work on gun control. Yet in those 40 years, she has not managed — bothered? — to learn basic terminology and concepts. This includes her prepared statements, indicating that she does have a staffer conversant on this supposed priority issue, nor has she bothered to even consult a semi-knowledgeable person when drafting those statements. You would have expected that Feinstein would have acquired a level of expertise simply as an unavoidable side-effect of listening to innumerable presentations and discussions over the many years. That she hasn’t has implications about what she has been listening to. Consequently, every time the Senate considers gun control, Feinstein makes the highlight reel of ignorant and befuddled statements.

The following was inspired by some of the comments on the Diana Diamond blog “^ Do something about assault weapons, now!^” (2019-08-13).

While this can be a very emotional issue, if you let your emotions dominate the discussion and treat your feelings as facts, are others going to think there is any point to interacting with you? This is a common enough situation to have dismissive phrases such as “Facts don’t care about your feelings” and “Facts over feelings”.

“I don’t need a gun, therefore no one needs a gun!” and its many variants.
“People don’t need guns!” says someone who thinks everyone is/should be just like them.

If your argument is nothing but a few slogans, such a low level of effort demonstrates that you are uninterested in trying to find a solution. Similarly, demonizing people with different perspectives signals that you are only interested in posturing, and add to the perception that your “tribe” will not act in good faith.

Example: “We need to ban automatic weapons.”
Uh. They have been banned for 85 years: by the National Firearms Act of 1934.
Or more generally, propose “solutions” that have already been implemented, or tried and failed.

Example: “No one needs an assault weapon for home defense. A pistol is enough.”
Effective use of a pistol for home defense — beyond scaring someone off — requires much more training and ongoing practice than is likely for most owners of home defense weapons. Plus being effective on the firing range is very different from being effective in high-stress real-world situations. Consider the typical shoot-out involving the police: The substantial majority of their bullets miss. Also, many pistol wounds are not only survivable, but not incapacitating in the short term. In contrast, a rifle is easier to control because it is held at 3 points — each hand and the shoulder — and can have less recoil (low recoil is often given as a major advantage of the AR-15). On the other hand, a rifle is much harder to store safely as a home defense weapon.

Example slogan: “Assault weapons are designed to kill many people quickly.”
Wrong. For the modern battlefield, soldiers are intensively trained to avoid bunching up and becoming vulnerable in this way.(foot#9) Yes, “assault weapons” are capable of quickly killing many people who are close together — just as a car is — but the slogan says designed. Word choices matter. Is the slogan signaling ignorance or an unwillingness to discuss in good faith?

Example slogan: “Assault weapons are weapons of war.” Disingenuous and only half-true. The term “weapons of war” can be stretched to such extremes as to be useless. Airplanes have been used as “weapons of war”. Should we ban them? Wine and liquor bottles were used for ^ Molotov Cocktails^ — should they be banned? One of the motivations for the Second Amendment to the US Constitution (right to bear arms) was that some states didn’t have the funds to arm their militias and expected many of those called up to bring their own personal (non-military) firearms. Designs developed and proven by the military routinely are adapted for non-military uses, but various features may be absent and ruggedness and reliability may be reduced. That is, the military versions are over-engineered and too expensive for most of the non-military market. However, if the need arises, these guns can be used as (inferior) substitutes for military weapons.

—-Conclusion—-

Above I have asked many questions that I — from my training as a research scientist — would expect to be asked and researched as part of developing and debating a policy. I hope that the questions I presented have not only encouraged you to think about which questions you regard as important. Hopefully, my questions triggered you to add ones of your own.

Of the questions that you think are important to the issue, how many of them have you heard addressed in the public debate? While there are some people asking these and similar questions, they are largely being ignored or shunted aside.

Remember that this is not a new issue, but the subject of a multi-decade debate. Ask yourself what the unasked and unanswered questions imply about the current iteration of the debate.

Finally, I reiterate that the sort of mass shootings driving the current moral panic are very rare, and that our intuitions about rare events can be very, very wrong.

—-Footnotes—-
1. Cautionary example of wrong intuitions about rare events:
Section “Base Rate Fallacy” of my blog ^ Swastikas, censorship, false positives and kittens^, 2017-09-07.

2. Cosmetic, not functional features:
Example: A stock that had a three-inch adjustment range to accommodate different size people was halfway to categorizing that rifle as an “assault weapon”. But not a rifle where end-caps are added to the stock to achieve the desired length. There are very legitimate concerns about stocks that allow a rifle to be significantly shorter (approx 9 inches?), and thus more concealable. However, the primary target of the ban was the basic AR-15, and it had the 3-inch adjustable stock. Is it even faintly plausible that those drafting the law were unaware of this? That including adjustable stocks was merely an oversight?
Example:Although a bayonet hasn’t been used in these mass shootings, having a bayonet mount similarly counted as 1 of 2 points needed to bean “assault weapon”. Why would a civilian rifle have such a mount? Shareable parts. Rather than a manufacturer having separate production lines for barrels for its military and non-military variants, it was likely cheaper to just leave a vestigial mount on the latter.
Example:In discussions, the AR-15 with a black plastic stock was seen as threatening and dangerous, whereas a functionally equivalent model with a “friendly” wood stock wasn’t.

3. AR-15 rifles widely available for 40 years:
The original patents expired in 1977, resulting in less costly variants.

4. Semi-automatic rifles widely available since at least the 1960s
A semi-automatic rifle — ^ M1 Garand^ — was the standard-issue rifle for the US military in WW2. I remember that semi-automatic designs were slow to enter the civilian market, but am not inclined to research it.
As a teen in the 1960s, I was shooting some semi-automatic rifles, mostly the .22-caliber ^ AR-7 Survival/Backpacker/Explorer^. Bolt-action rifles were preferred because of lesser problems with people for whom gun safety and especially muzzle discipline were still not a reflex. Aside: Shooting your hunting buddy in the face — ^ as then-VP Dick Cheney did^– is an example of failed muzzle discipline.
This was a time when the Vietnam War was escalating and boys some years older than me were already in combat there. A classmate had all three of her brothers become casualties on the same night.
Fathers who were combat veterans of WW2 or Korea were encouraging their sons to get head starts on various combat skills without referring to Vietnam. There was a decided difference between the fathers who had been in combat and those who hadn’t, with the former not romanticizing war. Aside: My father served first in the South Pacific (artillery) and then as a combat engineer at the very front of Patton’s Third Army. The latter involved darting through German lines to build a bridge that was expected to be ready by the time the tanks arrived, and to seize and secure the area until then.

5. 140 years of widely available rapid-fire rifles with large magazines:
These were first used in small quantities in the US Civil War — the ^ Henry Repeating Rifle^ — followed by the iconic ^ Winchester Lever-action Repeating Rifle, model 1873^, aka “The Gun that Won the West”, and its successors.
A 15-round magazine was typical.

6. Senseless shooting in 1970’s Detroit examples:
– There were people fatally shooting family members believing them to be intruders.
– There was the mother who was having a heated argument with her son who then stomped off, ignore her demands that he come back. She reached into a kitchen drawer for a gun and fatally shot him in the back.
– Down the street from a friend, there was a homeowner who was upset about his neighbor’s dog pooping on his lawn. He fired into the ground near the dog to scare it. The dog’s owner objected to this by threatening the shooter with his gun. Gunfire ensured, but I don’t remember the outcome.
– A couple had broken up and friends were helping one move out. A dispute arose over who got to keep the TV, becoming a shooting with one of the helpers killed.

7. Most deadly mass shootings for their time.
– The ^ Texas Tower shooting^ of 1966 is often used as the beginning of the modern era of mass shootings.
Killed: 13
Weapons: bolt-action rifle, pump-action rifle, semi-automatic carbine (not an “assault weapon”), semi-automatic shotgun, double-action revolver, two semi-automatic pistols. The revolver meets the definition of a semi-automatic, but is usually excluded from that classification.
– ^ San Ysidro McDonald’s, 1984^.
Killed: 21
Weapons: pump-action shotgun, semi-automatic carbine, semi-automatic pistol.
– ^ Luby’s, 1991^.
Killed: 23
Weapons: semi-automatic pistols.
– ^ Virginia Tech, 2007^.
Killed: 32
Weapons: semi-automatic pistols.
– ^ Orlando Nightclub, 2016^.
Killed: 49 (many of the deaths resulted from the failure of the police response)
Weapons: semi-automatic rifle and semi-automatic pistol.

Landmark school shootings:
– ^ Orlean NY, 1974^ (near where I grew up).
Killed: 3 (the shooting was from inside the school which was not in session, with one killed inside as he investigated)
Weapons: bolt-action rifle, shotgun.
– ^ Columbine, 1999^.
Killed: 13
Weapons: semi-automatic carbine using pistol ammunition, semi-automatic pistol with a large magazine, pump-action shotgun, double-barrel shotgun, 99 bombs that failed to explode.
– ^ 2012 Sandy Hook^
Killed: 28
Weapons: semi-automatic rifle (AR-15 style), semi-automatic pistol.
– ^ 2018 Parkland Florida^.
Killed: 17
Weapon: semi-automatic rifle (AR-15 style).

8.News article: “^ Third-grader pulls trigger on Maplewood cop’s gun, firing a shot^”, TwinCities.com/Pioneer Press, 2018-02-05.

9. Kill many people quickly:
I have seen from various calculations for various wars, and none of them find a ratio of bullets fired to soldiers killed of less than 10,000:1, and many are significantly higher. What was realized after WW2 is that most small-arms fire was used to suppress the enemy, that is, to encourage him to stay in cover rather than shooting back, and to pin him down while your unit maneuvers to advantageous positions.
This realization caused the US to switch to the 5.56mm round for the M16 from the 7.62mm round used by its predecessor — a weight reduction that allowed soldiers to carry more ammunition. The Soviets made a similar move when they replaced the AK-47/AKM assault rifle with the AK-74.

—-
An ^ abbreviated index by topic and chronologically^ is available.

—-Boilerplate on Commenting—-
The ^ Guidelines^ for comments on this blog are different from those on Town Square Forums. I am attempting to foster more civility and substantive comments by deleting violations of the guidelines.

I am particularly strict about misrepresenting what others have said (me or other commenters). If I judge your comment as likely to provoke a response of “That is not what was said”, do not be surprised to have it deleted. My primary goal is to avoid unnecessary and undesirable back-and-forth, but such misrepresentations also indicate that the author is unwilling/unable to participate in a meaningful, respectful conversation on the topic.
A slur is not an argument. Neither are other forms of vilification of other participants.

If you behave like a ^ Troll^, do not waste your time protesting when you get treated like one.